Why are property titles the reconstruction bottleneck?

Rebuilding Syria's housing is estimated at $75 billion — the biggest single component of the reconstruction bill and the one with no commercial buyer, no assigned donor and no announced state programme. Capital will not enter without security of title, and Syrian title is compromised by destroyed records, wartime displacement and expropriation, informal pre-war construction, and inheritance claims scattered across a diaspora of millions. Every housing finance instrument anyone has proposed — diaspora bonds, mortgage markets, pooled property funds — assumes a clean registry that does not currently exist. Fixing the registry is slower and less photogenic than pouring concrete, and nothing else starts until it moves.

Why property titles are the reconstruction bottleneck

Start with the number that frames everything. The World Bank’s October 2025 assessment puts reconstruction of Syria’s damaged physical assets at roughly $216 billion, against recorded direct damage of $108 billion. Break that down by asset class and residential buildings account for $75 billion of reconstruction against $33 billion of recorded damage — a multiplier of 2.3×, the widest gap of any category.

Now the part that is not in the arithmetic. Of the $25.9 billion in investment memoranda Syria has signed since December 2024, essentially none targets housing. The signed list is ports, power, telecoms, aviation, a metro. Look for the housing entry and you find, on our own Project Tracker, a single named residential development.

The usual explanation is that housing lacks a revenue stream. A port charges fees; an apartment block generates rent from a household that may be spending twice the minimum wage on food. That explanation is true and it is not the binding constraint.

The binding constraint is that you cannot lend against, insure, securitise or sell an asset whose ownership cannot be established.

What “compromised title” actually means here

Four distinct problems, which get lumped together and shouldn’t be, because they have different fixes.

Destroyed and inaccessible records. Land registries are physical archives. Some were destroyed with the buildings that housed them; some are intact but in areas whose administrative control changed repeatedly; some exist only on paper with no digital backup. A missing registry entry is not a legal dispute — it is an absence of evidence, and it defaults to paralysis.

Displacement without documentation. Roughly 1.6 million refugees and 1.9 million internally displaced people have returned since December 2024, per UNHCR as of May 2026. A large share left in circumstances that did not include collecting deeds. Proving a claim to a property you lived in for thirty years is straightforward when the registry confirms it and near-impossible when it does not and you have no paper.

Wartime expropriation and reallocation. Property was seized, reassigned, and in places redeveloped under legal instruments issued by the former government. Some of those transfers were routine, some were punitive, and distinguishing them case by case is a judicial task measured in years.

Pre-war informality. This one predates the conflict entirely and is routinely forgotten. A substantial share of Syrian urban housing was built informally, without permits, on land held under customary or undocumented arrangements. Those homes were real, occupied, and bought and sold — but they never had the kind of title a mortgage lender recognises. Rebuilding them to current codes converts an informal asset into a formal one, and somebody has to decide who the formal owner is.

Why capital treats this as disqualifying

Consider what a housing finance instrument actually requires.

A mortgage requires that the lender can take the property if the borrower defaults. That requires a registry entry the courts will enforce, and an enforcement process that completes in a predictable time. Neither currently exists at scale.

A diaspora property fund — the instrument most often proposed, and a genuinely good idea — requires that an investor in Berlin can be certain the fund owns what it says it owns. Fund managers will not accept title risk they cannot quantify, and their investors will not accept it at any price.

Construction lending requires that the site is not subject to a claim that surfaces halfway through the build. A contested title discovered at month eighteen is a total loss, not a delay.

Even outright grant-funded reconstruction hits it. A donor rebuilding a destroyed block needs to know who to hand the keys to. Hand them to the wrong claimant and the donor has funded an expropriation.

This is why the money went to ports. Not because investors are indifferent to housing, but because a port concession has one counterparty — the state — and a clear legal instrument. A residential block has as many counterparties as it has flats, and possibly more.

The sequencing problem

Here is what makes this urgent rather than merely difficult.

Returns are running ahead of reconstruction. Three and a half million people have come back to a housing stock that has not been rebuilt, and the World Bank’s April 2026 outlook is careful about it: return movements “are expected to create short-term pressures on already constrained public services and humanitarian assistance,” even as returnees eventually “support growth through renewed economic activity, increased labor supply, and the return of human and financial capital.”

Both halves are true. Which dominates depends on whether housing gets financed. And housing cannot get financed until title is resolvable.

Meanwhile every month of delay makes resolution harder, because people are moving into buildings, rebuilding informally, and establishing possession that will itself become a claim. Informal reconstruction is rational for a family with nowhere to live. It is also actively degrading the registry problem while everyone waits for a mechanism.

What a fix looks like

Nobody has to invent this from scratch. Post-conflict property restitution is a well-developed field, and the components are known.

A claims process with a deadline and an appeal. Bosnia’s Commission for Real Property Claims is the standard reference: an administrative body that adjudicates claims on documentary and testimonial evidence, faster than courts and with judicial review available. It is imperfect and it is enormously better than nothing.

Reconstruction of the registry from surviving evidence. Tax records, utility connections, court files, satellite imagery, and neighbour testimony can rebuild a defensible picture where the original record is gone. This is slow, unglamorous administrative work.

An explicit rule on wartime transfers. Not a case-by-case judicial slog but a stated legal position on which categories of transfer stand and which are void, so that the courts handle the exceptions rather than every file.

Recognition of informal pre-war holdings. A route by which an undocumented but genuine long-standing occupancy becomes formal title. Without this, formalising the registry dispossesses exactly the households least able to contest it.

Publication. A registry nobody can search is not much better than no registry. Searchability is what turns a legal fact into a financeable one.

What to watch

Three indicators, in order of how much they tell you.

Whether a claims mechanism is established at all, with a stated evidentiary standard and a published timetable. Everything else is downstream of this.
Whether wartime transfers get a general rule rather than case-by-case litigation.
The first mortgage. Not a housing MoU — an actual loan secured on Syrian residential property, written by a bank that expects to be able to enforce it. That is the signal that title has become financeable, and until it happens the $75 billion has no route to being spent.

Reconstruction coverage tends to track announcements, because announcements are events and registries are not. But the registry is the variable. A country can sign an unlimited number of housing memoranda while the deeds are unresolvable, and none of them will produce a flat.

Sources

This is a standing explainer. We update it when the facts change rather than republishing it as news.

First published 27 July 2026